Adaptive and self-adaptive techniques for evolutionary forecasting applications set in dynamic and uncertain environments

Abstract

Evolutionary Computation techniques have proven their applicability for time series forecasting in a number of studies. However these studies, like those applying other techniques, have assumed a static environment, making them unsuitable for many real-world forecasting concerns which are characterized by uncertain environments and constantly-shifting conditions. This chapter summarizes the results of recent studies that investigate adaptive evolutionary techniques for time series forecasting in non-static environments and proposes a new, self-adaptive technique that addresses shortcomings seen from these studies. A theoretical analysis of the proposed technique’s efficacy in the presence of shifting conditions and noise is given.